Jump to content

The Death of Tape


Recommended Posts

  • Premium Member

People are talking about the death of film but what's going to happen in 20 years to everyone's digital movies? People aren't even shooting as much tape anymore. It's all hard drives and mine barely last 5 years. Restore that!

 

Right now, LTO computer tapes are used for digital archiving of movies... and the idea (or ideal) is to back them up / copy them on a regular basis. Not a great plan, I agree... of course, bigger movies are recorded to 35mm b&w separations as well as stored digitally.

Link to comment
Share on other sites

  • Premium Member

People aren't even shooting as much tape anymore. It's all hard drives and mine barely last 5 years. Restore that!

 

Not that Tape has a stellar reputation for longevity. http://en.wikipedia.org/wiki/Tape_baking

 

Data center folks set up arrays of disks in a RAID configuration, and use the network to copy the data to more than one data center. Tape backups are cycled through, so that it is uncommon to actually have to refer to the backup tapes.

 

Of course we all know taht "Minor damage" to film requires a few frames to be re-touched. Minor damage to digital turns it into a pixel playground.

Link to comment
Share on other sites

My brother works in the archives department at Penn State University which deals A LOT with archival material (mostly sports). Hands down, he recommends FILM as the best way to preserve images. It lasts longer and IF there was a problem with a "viewing device," you can still hold film up to light and see pictures. You can't do that with videotape or a hard drive.

 

I still have 8mm film that my grandparents shot in the 1960s that we enjoy threading up. But I have some VHS and even 1" that takes great pains to see today.

Link to comment
Share on other sites

Yeah, curious if anyone is doing the film thing, that's a smaller organization... haha

 

I was thinking about investing into an LTO system, but the film thing really makes sense... except for the fact that I've seen prices around $320-450 a minute. yowzas.

Link to comment
Share on other sites

B&W film on polyester base, stored in proper conditions is about the most stable archival medium you can get for moving pictures. The images are recorded on special Separation stock, exposing each color frame as a red, green and blue record on B&W stock. To recover the information later the individual Red, green and blue information must be scanned and recombined in a single RGB file format.

Link to comment
Share on other sites

A couple of papers on this topic make fascinating reading. They are "The Digital Dilemma" and the follow up paper "The Digital Dilemma II". Excellent material for those interested.

 

What I find interesting is that those who are proclaiming the "death of film" seem never to comment on this issue. Neither do the manufacturers of video equipment.

 

Also, it would appear that those who advocate video on the grounds that working with film is much more expensive than working with video, never seem to have regard to the long term costs of archiving their work. Add these costs into the calculations, especially if one is talking of separation and long term storage on film, and I am far from convinced.

Link to comment
Share on other sites

I think what would put most people off archiving onto film nowadays is having or paying for the facilities to transfer the content to film and back again, along with physical storage space, demanding storage conditions, along with the time it takes and any loss of quality over time and/or through transferring content to the optical domain and back. Let alone the fear that manufacturers may not support the technology in 20 years time (ie. Try getting spare parts.)

Link to comment
Share on other sites

I think what would put most people off archiving onto film nowadays is having or paying for the facilities to transfer the content to film and back again, along with physical storage space, demanding storage conditions, along with the time it takes and any loss of quality over time and/or through transferring content to the optical domain and back. Let alone the fear that manufacturers may not support the technology in 20 years time (ie. Try getting spare parts.)

 

There is so much valuable media stored in film archives and in terms on long term storage, it's a much safer option than digital file formats which seem to have an extremely short life span. The best bet seems to be B & W colour separation. Storage is simpler than digital, because it doesn't need an active process of archiving compared to the digital media. The parts used on film equipment are mechanical, so easily manufactured by machine shops. Decoding obsolete computer files could be more difficult in 50 years.

 

It probably becomes more worthwhile on projects that have a high investment and a long economic life, such as feature films.

 

Unfortunately, long term archiving is the elephant in the room with the digital media. There have been a number of reports on the issue in recent years.

Link to comment
Share on other sites

There is so much valuable media stored in film archives and in terms on long term storage, it's a much safer option than digital file formats which seem to have an extremely short life span. The best bet seems to be B & W colour separation. Storage is simpler than digital, because it doesn't need an active process of archiving compared to the digital media. The parts used on film equipment are mechanical, so easily manufactured by machine shops. Decoding obsolete computer files could be more difficult in 50 years.

 

It probably becomes more worthwhile on projects that have a high investment and a long economic life, such as feature films.

 

Unfortunately, long term archiving is the elephant in the room with the digital media. There have been a number of reports on the issue in recent years.

I can see the value in that film is able to retain the quality of its pictures for many years, in comparison to hard drives for instance in which from my experience last around 5 years (normally because of mechanical failure - but not that SSD's are much better.) I think you're right in that it's more worthwhile for material with high economic life, I think the process of transferring to film and back isn't as flexible as accessing data held on a RAID array, although in the context of archival I don't think flexibility is necessarily an important factor, and when you're Twentieth Century Fox dealing with the original prints from 'Star Wars', I don't think the cost of doing so is an issue.

Link to comment
Share on other sites

  • Premium Member

Just a couple of factoids that are relevant to this:

 

- Film only lasts a really long time if it's kept in quite careful conditions, with controlled temperature and humidity. These things help tape too, but it isn't nearly so critical. Film can be a good archival medium, I think it's a stretch to assume it is by default.

 

- Since it hasn't been around for a century, we don't really know how tape will do in the long term; it's anticipated to be about as good.

 

- Some of the concerns of helical scan tape arrangements like DAT and most video tape formats do not apply to things like LTO which are intended for archive.

 

- No matter what you do to magnetic tape, the gradual rotation of the planet's magnetic field will gradually compromise high frequencies, over decades. Most tape formats are designed to minimise the effect this will have, but it's unavoidable. Tape also suffers from print-through, depending on the magnetic permeability of the base.

 

- Digital information should be stored uncompressed where possible because the effects of damage are then limited to the actual damaged areas, and don't propagate across neighbouring frames.

 

I'm not sure LTO is such a bad bet really.

 

P

Link to comment
Share on other sites

Just a couple of factoids that are relevant to this:

 

- Film only lasts a really long time if it's kept in quite careful conditions, with controlled temperature and humidity. These things help tape too, but it isn't nearly so critical. Film can be a good archival medium, I think it's a stretch to assume it is by default.

 

- Since it hasn't been around for a century, we don't really know how tape will do in the long term; it's anticipated to be about as good.

 

 

 

 

There are numerous examples of what is historically very important film which has survived for over 100 years. Examples are first world war footage and, at this particular time, film of the Titanic. I doubt very much whether that footage was archived in any sophisticated way for many of the years it has survived, indeed much of the German wartime film survived the destruction of Berlin and much of Germany in the run-up to the end of the second world war.

 

A comparison is how well video imagery has survived since it was developed. Even now, there is a huge question mark over its longevity and that follows a number of stages (if the process can be regarded as a single line of development, which it is possible to argue it is not) which still require a very considerable degree of "anticipation" or hope. Trying to view early video is almost impossible, and even where early imagery has survived, the quality is quite poor.

 

It is possible that tape will help with the longevity issue of video imagery, but we "know" that film works. It would be pretty disasterous to rely on a "possibility" and then find that the "possibility" did not deliver. That is not to say that it should not be tried out, but a sensible approach would be to continue with archiving on film whilst allowing tape or some other form of archiving to prove itself.

 

"Better the devil you know, than one whom you do not", comes to mind.

Link to comment
Share on other sites

There are numerous examples of what is historically very important film which has survived for over 100 years. Examples are first world war footage and, at this particular time, film of the Titanic. I doubt very much whether that footage was archived in any sophisticated way for many of the years it has survived, indeed much of the German wartime film survived the destruction of Berlin and much of Germany in the run-up to the end of the second world war.

 

A comparison is how well video imagery has survived since it was developed. Even now, there is a huge question mark over its longevity and that follows a number of stages (if the process can be regarded as a single line of development, which it is possible to argue it is not) which still require a very considerable degree of "anticipation" or hope. Trying to view early video is almost impossible, and even where early imagery has survived, the quality is quite poor.

 

It is possible that tape will help with the longevity issue of video imagery, but we "know" that film works. It would be pretty disasterous to rely on a "possibility" and then find that the "possibility" did not deliver. That is not to say that it should not be tried out, but a sensible approach would be to continue with archiving on film whilst allowing tape or some other form of archiving to prove itself.

 

"Better the devil you know, than one whom you do not", comes to mind.

I'd still be interested in viewing difference mattes between film before archival and after re-scan.

 

No one stores high value content on one LTO tape, nor one hard disk, so making comparisons between one LTO tape or one hard disk VS a roll of film isn't a practical comparison. Data redundancy by today's standards makes digital archival achievable theoretically for as long as we want it to.

 

What attracts me at least to digital archival over film is having the confidence that I'll get back exactly what I put in. Which is why I'd be interested in seeing PQA results between pre-archived material stored on film and then after archival and re-scan. You could probably argue that if quality was lost it was due to the re-scan not the negative, an optical to digital process I would assume will improve in years to come (a process improving faster than the deterioration of the negative.)

 

Admittedly I have very little experience with film, but I remain unsure about its ability to retain picture detail within such discrete quantisation levels found in digital systems, for instance 14-bit scanners - from archival through till re-scan.

Link to comment
Share on other sites

  • Premium Member

Admittedly I have very little experience with film, but I remain unsure about its ability to retain picture detail within such discrete quantisation levels found in digital systems, for instance 14-bit scanners - from archival through till re-scan.

 

 

That is the remarkable thing about film, it has no Quantization by itself. Which is why when a film with a damaged negative is "restored" to digital, most filmmakers would be sure to save both the damaged original, as well as a new 3 strip separation. SO that when the next cycle of technical obsolescence comes around they can decide if the best material is the pre or post digitaly processed version.

Link to comment
Share on other sites

That is the remarkable thing about film, it has no Quantization by itself. Which is why when a film with a damaged negative is "restored" to digital, most filmmakers would be sure to save both the damaged original, as well as a new 3 strip separation. SO that when the next cycle of technical obsolescence comes around they can decide if the best material is the pre or post digitaly processed version.

I can understand keeping the original negative, when scanning methods improve there may be value in re-scanning the original negatives and restoring them, as in the case of the 'Jaws' video posted in this thread.

 

However to scan the film, re-edit/touch and then burn back to film, and then re-scan, it's the difference information I'd be interested to see before and after this process. Is any quality lost during the transfer to film and back/is the film format able to retain picture detail with the same strict quantisation levels found in digital.

 

I'm not referring to quantisation in the sense that film has binary values, but in the sense of, is the format able to retain picture detail to a finer accuracy than say for instance, a 14-bit scanning system, how much does it drift, how linear in nature is the loss (if any.)

 

Like I said, I'm not saying film does drift or isn't accurate enough, but I'd still be interested in seeing test results before and after.

Link to comment
Share on other sites

Tape or film?

 

I haven't recorded anything with tape for about 3 years. Since digital became cheaper and had better quality

 

Digital can be recorded on tape as well, all the way up to HDCAM SR, I assume you mean using IT type recording media like flash memory and hard drives.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...